Applied Scientist Ii, in Sps Tech

Amazon Amazon · Big Tech · IN, KA, Bengaluru · Machine Learning Science

Applied Scientist II role at Amazon's Marketplace business, focusing on building and improving ML models to enhance seller experience, selection quality, and revenue. The role involves research, design, implementation, testing, and deployment of science solutions, collaborating with cross-functional teams. Requires expertise in machine learning, operations research, and statistics, with experience in feature engineering, modeling, and scalable inference on large datasets.

What you'd actually do

  1. Design, implement, test, deploy, and maintain innovative science solutions to accelerate our business.
  2. Create experiments and prototype implementations of new learning algorithms and prediction techniques
  3. Collaborate with scientists, engineers, product managers, and stakeholders to design and implement software solutions for science problems.
  4. Use best practices to ensure a high standard of quality for all of the team deliverables

Skills

Required

  • machine learning
  • operations research
  • statistics
  • feature engineering
  • modeling
  • probabilistic modeling
  • hyper-parameter tuning
  • scalable inference methods
  • latent variable models
  • algorithms and data structures
  • parsing
  • numerical optimization
  • data mining
  • parallel and distributed computing
  • high-performance computing
  • Java
  • C++
  • Python
  • professional software development

Nice to have

  • PhD
  • architectural concepts
  • algorithms
  • schedule tradeoffs
  • new opportunities

What the JD emphasized

  • strong analytical skills
  • practical experience
  • accelerate the business and make it profitable
  • directly impact Amazon’s Selection quality and maximize fee revenue
  • highly collaborative environment
  • understand the business requirements and translate them into complex analytical outputs
  • design tests to explain performance of the models from impact on customer and cost perspective
  • create ML models to capture features impacting performance
  • comfortable building prototypes, testing and improving them given the feedback from the real time data
  • able to present your model and findings to a various range of stakeholders
  • expertise in machine learning, operations research, and statistics
  • expertise in applying theoretical models in an applied environment relying heavily on the latest advances in machine learning, optimization, stochastic modeling, and engineering
  • feature engineering, modeling, probabilistic modeling, hyper-parameter tuning, scalable inference methods and latent variable models
  • dealing with very large data sets and requirements on throughput
  • Master's degree in engineering, technology, computer science, machine learning, robotics, operations research, statistics, mathematics or equivalent quantitative field
  • 3+ years of building machine learning models or developing algorithms for business application experience
  • 3+ years of solving business problems through machine learning, data mining and statistical algorithms experience
  • Experience in any of the following areas: algorithms and data structures, parsing, numerical optimization, data mining, parallel and distributed computing, high-performance computing
  • Experience in patents or publications at top-tier peer-reviewed conferences or journals
  • Experience programming in Java, C++, Python or related language
  • Experience in professional software development
  • Experience implementing algorithms using toolkits and self-developed code

Other signals

  • improve on the models that will directly impact Amazon’s Selection quality and maximize fee revenue
  • create ML models to capture features impacting performance
  • building prototypes, testing and improving them given the feedback from the real time data
  • expertise in applying theoretical models in an applied environment relying heavily on the latest advances in machine learning, optimization, stochastic modeling, and engineering
  • Challenges will involve dealing with very large data sets and requirements on throughput